METHOD AND CULTURE MEDIUM FOR EX VIVO CULTURING OF EPIDERMIS-DERIVED STEM CELLS

Abstract

The present invention relates to a method for culturing epidermis-derived stem cells comprising the step of culturing epidermis-derived stem cells in the presence of a three-dimensional extracellular matrix (3D-ECM) and a basal cell culture medium comprising: Epidermal Growth Factor (EGF); and/or a Vascular Endothelial Growth Factor (VEGF); and/or a Fibroblast Growth Factor (FGF); and further a ROCK (Rho-kinase) inhibitor. The present invention further relates to a method for ex vivo de novo generation of epidermis-derived stem cells. Furthermore, the present invention relates to an epidermis-derived stem cell that is obtainable by a method according to the present invention. Uses of said epidermis-derived stem cell, e.g. uses of said epidermis-derived stem cell for in vitro tissue production, in vitro drug discovery screening and medical applications, are also provided herein. The present invention further relates to a cell culture medium that is employed in the context of a method of the present invention.

Claims

1. A method for culturing epidermis-derived stem cells comprising the step of culturing epidermis-derived stem cells in the presence of a three-dimensional extracellular matrix (3D-ECM) and a basal cell culture medium comprising: Epidermal Growth Factor (EGF); and/or a Vascular Endothelial Growth Factor (VEGF); and/or a Fibroblast Growth Factor (FGF); and further a Rho-kinase (ROCK) inhibitor.

2. The method of claim 1, wherein said basal cell culture medium comprises said EGF, said VEGF, said FGF and said ROCK inhibitor.

3. The method of claim 1 or 2, wherein said Vascular Endothelial Growth Factor (VEGF) is selected from the group consisting of VEGF-164, VEGF-165, VEGF-120 and VEGF-121.

4. The method of any one of claims 1 to 3, wherein said Fibroblast Growth Factor (FGF) is selected from the group consisting of FGF-2, FGF-7, FGF-10 and FGF-18.

5. The method of any one of claims 1 to 4, wherein said ROCK inhibitor is selected from the group consisting of (S)-(+)-2-methyl-1-[(4-methyl-5-isoquinolinyl)sulfonyl]-hexahydro-1H-1,4 diazepine dihydrochloride (H-1152) and (R)-(+)-trans-4-(1-aminoethyl)-N-(4-Pyridyl)cyclohexanecarboxamide dihydrochloride monohydrate (Y-27632).

6. The method of any one of claims 1 to 5, wherein said basal cell culture medium further comprises a Sonic Hedgehog (SHH) inhibitor.

7. The method of any one of claims 1 to 6, wherein said basal cell culture medium further comprises ethanolamine, phospho-ethanolamine and/or transferrin.

8. The method of any one of claims 1 to 7, wherein said epidermis-derived stem cells are hair follicle stem cells (HFSCs) or cancer stem cells (CSCs).

9. The method of any one of claims 1 to 8, wherein said epidermis-derived stem cells are comprised in a mixture of cell types, which further comprises at least one differentiated epidermal cell type.

10. The method of any one of claims 1 to 9, wherein said epidermis-derived stem cells are de novo generated ex vivo.

11. A method for ex vivo de novo generation of epidermis-derived stem cells comprising the step of culturing epidermal cells lacking said epidermis-derived stem cells in the presence of a three-dimensional extracellular matrix (3D-ECM) and a basal cell culture medium as defined in any one of claims 1 to 7 for at least 2 days, at least 4 days, at least 6 days, at least 8 days, at least 10 days, at least 12 days, at least 14 days, at least 16 days, at least 18 days, at least 20 days, at least 100 days, at least 200 days or at least 360 days.

12. An epidermis-derived stem cell that is obtainable by a method as defined in any one of the claims 1 to 11.

13. The epidermis-derived stem cell of claim 12 for use in tissue transplantation and/or for use in treatment of dermal burn, treatment of conditions where areas of skin have been removed due to surgical operation, biopsy, burn and/or trauma, and/or in treatment of conditions where the regenerative capacity of the skin is compromised such as chronic wounds and/or baldness.

14. Use of the epidermis-derived stem cell of claim 12 for in vitro tissue production and/or for in vitro drug discovery screening.

15. A cell culture medium as defined in any one of claims 1 to 7.

Description

[0348] The Figures show:

[0349] FIG. 1: Standard keratinocyte culture conditions do not support growth of CD34.sup.+6.sup.+ HFSCs.

[0350] FACS plots of freshly isolated mouse keratinocytes (day 0; d0) and keratinocytes cultured under standard 2D culture conditions for 2 weeks (day 14; d14). Single cell suspensions were stained for the stem cell markers 6 integrin and CD34. Gates were drawn according to the respective unstained and isotype-stained controls. Percentages are indicated per quadrant.

[0351] FIG. 2: Cell growth under different culture conditions

[0352] Representative bright field images of epidermal cells grown in the indicated growth media in the presence of 3D-ECM (scale bars 250 m). Y: EGF+Y27632; YV: Y27632+EGF+VEGF; YF: Y27632+EGF+FGF-2; 3C: Y27632+EGF+VEGF+FGF-2.

[0353] FIG. 3: CD34.sup.+6.sup.+ HFSC enrichment under different culture conditions

[0354] A. Percentage of CD34.sup.+6.sup.+ cells from day 14 (d14) HFSC cultures in different growth media. CD34.sup.+6.sup.+ cells were quantified by flow cytometry (meanSEM, n=4-5; *p0.05; **p0.01 Mann-Whitney U test). Epi d0: Freshly isolated epidermal cells, KGM 2D: basal medium in 2D; 3C 2D: Y27632+EGF+VEGF+FGF-2 in 2D; Y-E: Y27632 without EGF in the basal medium in 3D-ECM; Y: EGF+Y27632 in 3D-ECM, YV: Y27632+EGF+VEGF in 3D-ECM; YF: Y27632+EGF+FGF-2 in 3D-ECM; 3C: Y27632+EGF+VEGF+FGF-2 in 3D-ECM.

[0355] B. Absolute numbers of CD34.sup.+6.sup.+ cells from day 14 (d14) HFSC cultures in different growth media. CD34.sup.+6.sup.+ cells were quantified by flow cytometry. Data are plotted as fold enrichment over freshly isolated cells (meanSEM, n=4-5; *p0.05; **p0.01 Mann-Whitney U test). Y: Y27632+EGF; YV: Y27632+EGF+VEGF; YF: Y27632+EGF+FGF-2; 3C: Y27632+EGF+VEGF+FGF-2. All conditions are in 3D-ECM.

[0356] C. Normalized numbers of CD34.sup.+6.sup.+ cells from day 14 HFSC cultures in 3C media with (3C) and without (3C-E) EGF. CD34.sup.+6.sup.+ cells were quantified by flow cytometry. Data are normalized to 3C in each of the three independent experiments shown. SEM, n=3. 3C: Y27632+EGF+VEGF-164+FGF-2; 3C-E: Y27632+VEGF-164+FGF-2 without EGF in basal medium. All conditions are in 3D-ECM.

[0357] D. Representative FACS plots of freshly isolated (day 0; d0) and keratinocytes cultured in 3C conditions for 2 weeks (day 14; d14). Gates were drawn according to the respective unstained and isotype-stained controls. Percentages are indicated per quadrant.

[0358] E. Normalized percentages of CD34.sup.+6.sup.+ cells from day 14 HFSC cultures grown under 3C conditions for the indicated times. CD34.sup.+6.sup.+ cells were quantified by flow cytometry. For 3C-Y conditions cells were grown from day 0 to either day 2 or day 4 in 3C conditions and then the medium was exchanged to a 3C medium lacking Y27632 (3C-Y). Data are normalized to 3C conditions in each of the two independent experiments shown. SD, n=2. 3C: Y27632+VEGF-164+FGF-2; 3C-Y: VEGF-164+FGF-2; all in basal medium. All conditions are in 3D-ECM. d=days.

[0359] F. Normalized absolute numbers of CD34.sup.+6.sup.+ cells from day 14 HFSC cultures grown under 3C conditions for the indicated times. CD34.sup.+6.sup.+ cells were quantified by flow cytometry. For 3C-Y conditions cells were grown from day 0 to either day 2 or day 4 in 3C conditions and then the medium was exchanged to a 3C medium lacking Y27632 (3C-Y). Data are normalized to 3C conditions in each of the two independent experiments shown. SD, n=2. 3C: Y27632+VEGF-164+FGF-2; 3C-Y: VEGF-164+FGF-2; all in basal medium. All conditions are in 3D-ECM. d=days.

[0360] FIG. 4: CD34.sup.+6.sup.+ HFSC enrichment with different growth factor combinations

[0361] CD34.sup.+6.sup.+ cells were quantified by flow cytometry from day 14 Matrigel cultures containing various growth factor combinations. Normalized CD34.sup.+6.sup.+ content of cultures is shown. A. Direct comparison of different VEGF and FGF isoforms in 3C conditions (3D-ECM; basal cell culture medium comprising Y27632+EGF+FGF+VEGF). VEGF-121 performs equally well as VEGF-164 (set to 1). FGF-12, FGF-10, and FGF-7 perform equally well as FGF-2 (set to 1). B. Direct comparison of VEGF-121 and VEGF-164 in YV conditions (3D-ECM; basal cell culture medium comprising Y27632+EGF+VEGF). VEGF-121 performs slightly better than VEGF-164 (set to 1). C. Direct comparison of different FGF isoforms in YF conditions (3D-ECM; basal cell culture medium comprising Y27632+EGF+FGF). FGF-12, FGF-10, and FGF-7 perform equally well as FGF-2 (set to 1).

[0362] FIG. 5: CD34.sup.+6.sup.+ HFSC enrichment in 3D-ECMs with different compositions

[0363] CD34.sup.+6.sup.+ cells were quantified by flow cytometry from day 14 3C cultures in various ECM component-containing 3D-ECM gels. CD34.sup.+6.sup.+ content normalized to 3C Matrigel cultures is shown. Col I: Collagen type I; L332: laminin 332; L511: laminin 511; BME: basement membrane extract.

[0364] FIG. 6: 3C cultures enrich for CD200.sup.+ cells

[0365] A. Percentage of CD200.sup.+ cells in day 14 (d14) HFSC cultures. CD200.sup.+ cells were quantified by flow cytometry. Fold enrichment is shown over freshly isolated cells (meanSEM, n=3).

[0366] B. Representative FACS plots of freshly isolated (day0) and keratinocytes cultured in 3C conditions for 2 weeks (day 14; d14). Gates were drawn according to the respective unstained and isotype-stained controls. Percentage of CD200.sup.+ from CD34.sup.+6.sup.+ cells is indicated per quadrant.

[0367] FIG. 7: Morphology of cells cultured under 3C conditions

[0368] Representative image of cell clusters formed in 3C cultures of HFSCs. Note that no lumen is observed. BF: bright field; PH: phase contrast. Scale bars 100 m and 50 m (right and left panel, respectively).

[0369] FIG. 8: Proliferation of cells cultured under 3C conditions

[0370] A. FACS histograms of day 14 (d14) 3C cultures of HFSCs labeled with EdU for 24 h (day 13-day 14) and 48 h (day 12-day 14). Percentage of EdU+ cells in CD34.sup.+6.sup.+ and CD34.sup.6.sup.+ cells is indicated. Cells without EdU are shown as control. The Figure shows a single experiment.

[0371] B. FACS histograms of eight different 3C cultures of HFSCs labeled with EdU for 24 h (between days 10-11 and 12-13). Percentage of EdU+ cells in CD34.sup.+6.sup.+ and CD34.sup.6.sup.+ cells were quantified by flow cytometry. Cells without EdU are shown as control in black color.

[0372] C. Pooled results from individual experiments shown in panels A and B (meanSEM, n=8; *p0.05 Mann-Whitney U test)

[0373] FIG. 9: Expression of stem cell markers is maintained in 3C conditions

[0374] A and B. Confocal images of HFSCs grown under 3C conditions and stained for the indicated proteins. K15: Keratin-15. Scale bar 100 m.

[0375] FIG. 10: 3C cultures contain exclusively epithelial cells

[0376] Representative FACS histogram of day 14 (d14) HFSC cultures stained for EpCAM and quantified by flow cytometry. Unstained cells are shown as control. Percentage of EpCAM.sup.+ cells SD from 6 independent cultures is shown.

[0377] FIG. 11: Transcriptomes of cells grown in 3C cultures resemble transcriptomes of bona fide HFSCs

[0378] A. Principal component analysis of the transcriptome obtained from RNA sequencing experiments of i) epidermis-derived cells (Epi d0), ii) FACS-purified CD34.sup.+6.sup.+ HFSCs (CD34.sup.+6.sup.+), and iii) epidermis-derived cells cultured under 3C conditions for 14 days (3C). Purified CD34.sup.+6.sup.+ HFSCs and cells cultured in 3C conditions cluster together and are distinct from epidermis-derived cells indicating that their transcriptomes share significant similarity.

[0379] B. Principal component analysis of the transcriptome obtained from RNA sequencing experiments of i) epidermis-derived cells (Epi d0), ii) FACS-purified CD34.sup.+6.sup.+ HFSCs (CD34.sup.+6.sup.+), and iii) epidermis-derived cells cultured under 3C conditions for 14 days (3C). Plot shows individual biological replicates that cluster together indicating good reproducibility. As in panel A, purified CD34.sup.+6.sup.+ HFSCs and cells cultured in 3C conditions cluster together and are distinct from epidermis-derived cells indicating that their transcriptomes share significant similarity.

[0380] C. Heat map analysis of quantified transcripts and dendrogram show that the transcriptomes of cells cultured in 3C more closely resemble transcriptomes of CD34.sup.+6.sup.+ HFSCs than transcriptomes of the original cell mixtures (Epi d0). For each condition three biological replicates (1, 2 and 3) are shown. The represented data is derived from the same experiment as A.

[0381] D. Quantitative PCR (qPCR) analysis of Epi d0 and 3C shows that cells in 3C conditions upregulate HFSC identity genes (e.g. Cd34, Sox9, Tcf3, Wnt7a).

[0382] FIG. 12: Long-term passage of 3C cultures

[0383] A. Passage scheme. B. Percentage of CD34.sup.+6.sup.+ cells from HFSC cultures quantified by flow cytometry. Individual experiments are depicted as single points. SEM is shown.

[0384] FIG. 13: HFSCs cultured under 3C conditions can be freeze-thaw

[0385] Representative FACS plots of HFSC cultures before (day 14; d14) and 14 days after freeze-thaw. Cells were frozen for at least a month before thawing them. Thawed cells grew for additional 14 days before analysis.

[0386] FIG. 14: In vitro proliferative potential of HFSCs cultured under 3C conditions

[0387] Colony forming assays were performed with 3000 cells/well derived from three mice. Representative results (A) and quantification (B) are shown. Freshly isolated epidermal cells were used as a control (meanSEM; *p 0.05, Mann-Whitney U-test). YF: Y27632+EGF+FGF-2; 3C: Y27632+EGF+VEGF+FGF-2. All conditions were performed in triplicates.

[0388] FIG. 15: Morphology of colonies from colony forming assay of HFSCs cultured under 3C conditions

[0389] Representative images of characteristic colonies generated by the indicated cells. Scale bar 100 m. Scale bar 100 m. YF: Y27632+EGF+FGF-2; 3C: Y27632+EGF+VEGF+FGF-2.

[0390] FIG. 16: HFSCs cultured under 3C conditions maintain their multipotency and self-renewal potential in a skin reconstitution assay

[0391] A. Representative images of recipient mice transplanted with either freshly isolated epidermal cells (Epi d0) or cultured cells cultured under 3C conditions (3C). Note that animals transplanted with cells cultured in 3C developed more hair than Epi d0 controls indicating their multipotency and self-renewal capacity after 14 days of culture. B. H&E staining from transplant's biopsies. C. Hair follicle quantification from Hematoxylin & Eosin staining.

[0392] FIG. 17: De novo generation of HFSCs from non-HFSCs under 3C culture conditions

[0393] A. CD34.sup.+6.sup.+ HFSCs and CD34.sup.6.sup.+ non-HFSCs were purified from total epidermis into at least 98% purity and cultured under 3C conditions for 14 days. FACS plots of day 14 cultures established from the indicated purified cell populations show that both cultures consist of a significant population of CD34.sup.+6.sup.+ HFSCs and CD34.sup.6.sup.+ non-HFSCs.

[0394] B. CD34.sup.+6.sup.+ HFSCs were completely depleted by culturing total epidermal cells in 2D growth conditions for 14 days (2D 14 d). The resulting CD34.sup.6.sup.+ non-HFSCs were subsequently passaged and cultured under 3C conditions for additional 14 days (3D-3C d14). Shown are FACS blots of the indicated cells and the percentage of CD34.sup.+6.sup.+ HFSCs is indicated.

[0395] FIG. 18: 3C HFSC cultures reach equilibrium after 12 days of culture

[0396] CD34.sup.+6.sup.+ and CD34.sup.6.sup.+ cells in 3C cultures were monitored every 2 days for a period of 14 days. Cell populations (A) and absolute numbers (B) were quantified by flow cytometry.

[0397] FIG. 19: Inhibition of SHH signaling increases the number of CD34.sup.+6.sup.+ cells in 3C HFSC cultures

[0398] Cells were grown for 9 days under 3C conditions and treated with the SHH inhibitor cyclopamine (10 M) for 5 days. Treated cultures and untreated 3C control cultures were analyzed by flow cytometry on day 14.

[0399] FIG. 20: Inhibition of BMP signaling decreases the numbers of CD34.sup.+6.sup.+ cells in 3C HFSC cultures

[0400] Cells were grown for 14 days under 3C conditions with the BMP inhibitor dorsomorphin (2 M) (A) or the BMP inhibitor K02288 (10 nM) (B). Treated cultures and untreated 3C control cultures were analyzed by flow cytometry.

[0401] FIG. 21: Papillomas contain phenotypically defined CSCs in a mouse model of skin cancer

[0402] Papillomas arising in the K14rTA tet-O-Kras mouse model were stained for CSCs markers defined as Lin.sup.EpCAM.sup.+CD34.sup.+6.sup.+. Representative FACS plots from unaffected skin and tumor material. Gates were drawn according to the respective unstained and isotype-stained controls. Percentages are indicated per quadrant. n>12 tumors; n=5 mice. SEM is shown.

[0403] FIG. 22: Papilloma CSCs can be grown and maintained in vitro under different culture conditions

[0404] Single cell suspensions prepared from papillomas of the K14rTA tet-O-Kras mouse model were grown using different media that support HFSC growth. A. FACS plots showing CSC staining after 14 days of culture. Tumor: cells isolated from tumors; KGM: keratinocyte growth medium. YV: Y27632+EGF+VEGF; YF: Y27632+EGF+FGF-2; 3C: Y27632+EGF+VEGF+FGF-2.

[0405] FIG. 23: 3C culture conditions support enrichment of papilloma CSCs in long-term cultures

[0406] A. FACS plots of papilloma cells before (Tumor: freshly isolated tumor cells) and after (2 weeks of culture (p0) and 12 weeks of culture (p5)) culture. B. Percentage of EpCAM.sup.+CD34.sup.+6.sup.+ cells from papilloma CSC cultures quantified by flow cytometry.

[0407] FIG. 24: Morphology of papilloma CSCs cultured under 3C conditions

[0408] Representative bright field images of papilloma cells grown under 3C conditions for 12 weeks. 5 magnification. Scale bars 100 m (left) and 250 m (right).

[0409] The invention will now be described by reference to the following examples which are merely illustrative and are not to be construed as a limitation of the scope of the present invention.

EXAMPLE 1: DEVELOPMENT OF CULTURE CONDITIONS FOR CULTURING HFSCS AND EXPERIMENTAL CHARACTERIZATION OF SUCH CULTIVATED HFCSS

Materials and Methods

Mice

[0410] Keratinocytes were isolated from C57BL/6 mice on postnatal day (P) 21, unless stated otherwise. 7-9 weeks old BALB/c-nude mice (CAnN.Cg-Foxn1nu/Crl; Charles River, Germany) were used as recipients in transplantation experiments. Primary dermal fibroblasts were isolated from C57BL/6 mice on P2. Animals were housed and maintained according to FELASA guidelines in the animal facility of the Max Planck Institute for Biology of Ageing, Cologne, Germany. All experiments were approved by local authorities.

Reagents and Antibodies

[0411] The following components were used to prepare the basal keratinocyte growth medium (KGM basal medium): MEM (Spinners modification, Sigma), 5 g/mL Insulin (Sigma), 10 g/mL Transferrin (Sigma), 10 M Phosphoethanolamine (Sigma), 10 M Ethanolamine (Sigma), 0.36 g/mL Hydrocortisone (Calbiochem), 2 mM Glutamine (Gibco), 100 U/mL Penicillin and 100 g/mL Streptomycin (Gibco), 8% chelated fetal calf serum (FCS) or fetal bovine serum (FBS) (Gibco).

[0412] The following components were used to prepare the FAD medium: 2 parts of DMEM (Gibco) and 1 part of Ham's F12 (Gibco), 2 mM Glutamine (Gibco), 100 U/mL Penicillin and 100 g/mL Streptomycin (Gibco), 10% chelated FCS or FBS (Gibco), 50 g/ml Vitamine C (Sodium L-ascorbate; Sigma), 10 ng/ml EGF (Sigma), 5 g/ml Insulin (Sigma), 0.5 g/ml Hydrocortisone (Calbiochem, 0.1 nM cholera toxin (Sigma), 180 M Adenin (Sigma).

[0413] Growth factor reduced Matrigel (BD Biosciences), (R)-(+)-trans-4-(1-aminoethyl)-N-(4-Pyridyl)cyclohexanecarboxamide dihydrochloride monohydrate (referred to as Y-27632; Sigma-Aldrich), human recombinant FGF-2, human recombinant FGF-7, human recombinant FGF-10 (all from Miltenyi Biotec), human recombinant FGF-18 (Peprotech) mouse recombinant VEGF-164, human recombinant VEGF-121 (all from Miltenyi Biotec), human recombinant EGF (LONG-EGF; Sigma), rat tail collagen type I (Millipore), human recombinant Laminin-332 and human recombinant Laminin-511 (Biolamina) were used for cell culture experiments. Collagenase (C0130, Sigma) was used for cell recovery from collagen type I gels. 7AAD (eBioscience) and Fixable Viability Dye eFluor405 (eBioscience) were used to discriminate dead cells for flow cytometry.

[0414] The following antibodies were used for immunostaining: eFluor660-CD34 (clone RAM34, eBioscience), Pacific blue-alpha-6 Integrin (clone GoH3, eBioscience), FITC-CD133 (clone 13A4, eBioscience), PE-CD200 (clone OX90, eBioscience), PE-Cy7-EpCAM (clone G8.8, eBioscience), PE-CD140a (clone APA5, eBioscience); PE-CD45 (clone 30-F11, eBioscience), PE-CD31 (clone MEC13.3, eBioscience), APC-Cy7-beta-1 Integrin (clone HMB1-1, eBioscience), SOX9 (H-90, Santa Cruz Biotechnology), Keratin15 (LHK15, NeoMarkers, Fremon, Calif.). Secondary antibodies for immunofluorescence anti-mouse AlexaFluor 488 and anti-rabbit AlexaFluor 568 were from LifeTechnologies.

Cell Isolation and Culture

[0415] Keratinocytes were isolated from back skin of mice by incubating skin pieces in 0.8% Trypsin (Gibco) for 50 min at 37 C. After separating the epidermis from the underlying dermis, cells were homogenized, filtered through 70 m and 45 m cell strainers (BD Biosciences) and pelleted at 900 rpm for 3 min. For 2D culture of epidermis-derived cells, tissue culture petri dished were coated with a mixture of collagen I (30 g/ml)+fibronectin (10 g/ml; both from Millipore) in MEM for 1 h at 37 C. prior to plating of cells. Epidermis-derived cells were suspended in KGM or in FAD medium, and seeded on the coated plastic dishes. Medium was exchanged the next day after initial seeding and thereafter every second day. Cultures were incubated at 37 C., 5% CO.sub.2 for 12-14 days. For 3D culture of HFSCs, 810.sup.4 cells were resuspended in 40 l ice-cold 1:1 KGM:Matrigel mixture that was dispensed as a droplet in 24-well cell culture dishes. The suspension was allowed to solidify for 15 min after which it was overlaid with 500 l of stem cell media. The following media compositions were used: KGM basal medium, medium Y (KGM, 5 M Y27632, 10 ng/ml human recombinant EGF), medium Y-E (KGM, 5 M Y27632), medium YV (KGM, 5 M Y27632, 10 ng/ml human recombinant EGF, 20 ng/ml mouse recombinant VEGF), medium YF (KGM, 5 M Y27632, 10 ng/ml human recombinant EGF, 20 ng/ml human recombinant FGF-2), 3C (KGM, 5 M Y27632, 10 ng/ml human recombinant EGF, 20 ng/ml mouse recombinant VEGF, 20 ng/ml human recombinant FGF-2), and 3C-E (KGM, 5 M Y27632, 20 ng/ml mouse recombinant VEGF, 20 ng/ml human recombinant FGF-2). Cultures were incubated at 37 C., 5% CO.sub.2 for 12-14 days during which the medium was replaced every 2 days.

[0416] For cell passaging and flow cytometry cultured hair follicle stem cells were extracted from Matrigel by mechanical homogenization and incubation in 0.5% Trypsin (Gibco), 0.5 mM EDTA, PBS or Accutase (Gibco) for 10 min at 37 C. Cells were passaged every 10-14 days in a 1:3 to 1:5 ratio in fresh 1:1 KGM (where necessary with the respective supplements):Matrigel.

Proliferative Potential Assay

[0417] A colony-forming assay to determine proliferative potential of stem and progenitor cells was conducted as previously described (Jensen et al., 2010, Nat Protoc 5: 898-911). Briefly, 2000-4000 cells were plated on 6-well plates containing Mitomycin C-treated feeder cells (J2 fibroblasts). Cultures were incubated at 37 C., 5% CO.sub.2 for 12-14 days during which the medium was replaced every 2 days. Experiments were terminated when colonies reached a sufficient size to be visually identified and quantified. Colonies were fixed with 4% PFA 10 min and then stained with 1% crystal violet. Colony number and area were determined using the ImageJ software. All conditions were performed in triplicates.

Flow Cytometry (FACS Analysis and FACS Purification)

[0418] Single cell suspensions prepared from murine back skin (described above) or from cultured cells were rinsed once with KGM and stained with fluorescently labeled antibodies for 30 min on ice. After two washes with FACS Buffer (2% FCS, 2 mM EDTA, PBS) cells were measured in a BD FACS Canto II and data was analyzed using FlowJo software version 10. For purification of cell populations (cell sorting) a BD FACS Aria II or a BD FACS Aria Fusion were used. Cells were sorted in KGM medium at 4C. Expression of cell surface markers was analyzed on live cells after exclusion of cell doublets and dead cells, respectively.

EdU Incorporation Assay

[0419] Cells were grown in the presence or absence of 9.4 m EdU (LifeTechnologies) for 24 or 48 h before analysis. After preparing single-cell suspensions, cells were stained with a fixable viability dye eFluor506 (eBiosciences) followed by antibody staining before fixation in 4% PFA for 10 min at RT. Cells were next permeabilized in 0.1% Triton X-100, PBS for 10 min, and incubated 30 min in EdU reaction cocktail (100 mM Tris pH 8.5, 1 mM CuSO.sub.4, 0.5 M AlexaFluor-488-Azide (A10266, Life Technologies), 100 mM ascorbic acid). After 2 washes with 0.1% Triton X-100, PBS cells were analyzed by flow cytometry.

Hematoxylin and Eosin Staining

[0420] Paraffin-embedded tissue sections were deparaffinized with Xylol and rehydrated with consecutive washes of 100% isopropanol, 95%, 75%, 50% ethanol, and distilled water. Sections were stained for 1 min with Hematoxylin and counterstained for 10 sec with Eosin. Sections were then dehydrated in 50%, 75%, 95% ethanol and isopropanol, cleared with Xylol, and mounted in Entellan (Merk).

Immunofluorescence

[0421] Cultured hair follicle stem cells were rinsed once in PBS, followed by fixation in 2% PFA, PBS for 30 min at room temperature (RT). Fixed cells were rinsed three times with 100 mM glycine, PBS, then permeabilized and blocked for 2 h at 37 C. in 0.3% Triton-X 100, 5% BSA, PBS. Cells were stained with unlabeled or fluorescent primary antibodies in 0.3% Triton-X 100, 1% BSA, PBS overnight at RT. Secondary, fluorescent antibodies were used to detect primary antibody binding and nuclei were visualized with DAPI. Slides were mounted with Elvanol mounting medium (0.2 M Tris pH 6.5, 12% polyvinyl alcohol, 30% glycerol, 2.5% DABCO-anti fade reagent).

Microscopy and Image Analysis

[0422] Fluorescent images were collected by laser scanning confocal microscopy (TCS SP5X; Leica) with 63 or 40 immersion objectives using LAS X software. All images were recorded sequentially and averaged at least twice. Image processing (linear brightness and contrast enhancement) was performed with Fiji Software version 2.0.0 or Adobe Photoshop CSS.

Full Thickness Skin Reconstitution Assay (Transplantation Assay)

[0423] Transplantation of keratinocytes to assess their multipotency and self-renewal capacity was performed essentially as previously described (Blanpain et al., 2004, Cell 118: 635-648; Jensen et al., 2010, Nat Protoc 5: 898-911). Briefly, a mixture of 1-510.sup.5 freshly isolated keratinocytes or cultured cells together with 510.sup.6 freshly isolated neonatal fibroblasts was injected into a silicon chamber inserted into a full thickness wound on the back skin of nude mice (6-8 weeks old). The chamber was removed a week after transplantation and hair growth was monitored for the following 2-4 weeks.

Real Time Quantitative PCR (RT-qPCR)

[0424] RNA was extracted with the RNeasy Plus Mini Kit (Qiagen). cDNA was synthesized with the SuperScript VILO (LifeTechnologies) or the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems). qPCR was performed on the CFX384 Touch Real Time PCR Detection System (Bio-Rad) and the StepOne Plus Real Time PCR System (Applied Biosystems) with DyNAmo ColorFlash SYBR Green Mix (Thermo Fisher). Gene expression changes were calculated following normalization to an ERCC spike-in reference RNA (Ambion-LifeTechnologies) using the comparative Ct (cycle threshold) method.

RNA Sequencing

[0425] Total RNA was extracted as describe above for RT-qPCR. RNA quality was evaluated with an Agilent 2200 TapeStation. Three biological replicates/condition were sequenced. Libraries were made with NEBNext Ultra Directional RNA Library Prep Kit (New England Biolabs) followed by sequencing with HiSeq 2500 (Illumina).

[0426] Reads were mapped to the Mus musculus reference genome (build GRCm38_79), after quality control, followed by differential gene expression analysis using Cufflinks (version 2.2.1). Transcripts regulated 2 log 2 fold change and with adjusted p-value0.05 were considered significantly regulated. The principal component analysis (PCA) was performed using the cummeRbund package and R software. The data have been submitted to NCBI-GEO (GSE76779).

Statistical Analysis

[0427] Statistical analyses were performed using GraphPad Prism software (GraphPad, version 5.0). Statistical significance was determined by the Mann-Whitney U-test or ANOVA test.

Results

Development of Culture Conditions for Expansion and Maintenance Hair Follicle Stem Cells

[0428] Murine keratinocytes were cultured employing standard culture conditions. Standard growth conditions are defined by the use of KGM or FAD medium, a widely used standard keratinocyte culture medium (Watt & Green, 1982, Nature 295:434-436), as growth media. In addition the cells are cultured in 2-dimensional (2D) culture conditions, which are achieved by coating a tissue culture petri dish with ECM components (such as Collagen I), allowing them to polymerize and subsequently plating the cells on top of the ECM surface.

[0429] Flow cytometry analyses of murine keratinocytes cultivated under these conditions or freshly isolated murine keratinocytes for comparison, demonstrated that the CD34.sup.+6.sup.+ HFSC population was rapidly depleted (e.g. within 14 days) under said standard culture conditions (FIG. 1). Freshly isolated keratinocytes from a P21 mouse contained 5.61.2% (SD) CD34.sup.+6.sup.+ HFSCs (FIG. 1 and FIG. 3).

[0430] In order to develop improved in vitro culturing conditions for HFSCs, the role of laminins and collagens in a three-dimensional (3D) ECM microenvironment was assessed. To culture the cells in the presence of a 3D-ECM, single cell suspensions were embedded within the ECM mixture prior to polymerization of the ECM components, allowing the cells to be embedded within the ECM from all sides. The survival and growth/expansion of keratinocytes in 3D ECM gels composed of collagen type I, collagen type I and laminins, and a laminin-rich basement membrane extract (Matrigel (Corning) or Culturex (Amsbio)), was subsequently analyzed.

[0431] In particular, freshly isolated keratinocytes were embedded in Matrigel or Culturex (from now on referred to as 3D-Matrigel ECM), and cultured in KGM comprising EGF. Under these conditions no cell growth was observed. Yet, keratinocyte survival and/or growth/expansion could be established in the presence in 3D-Matrigel ECM by adding EGF and a ROCK inhibitor (e.g. Y27632) to the KGM culture medium (FIG. 2 and FIG. 3, see Y). As determined by flow cytometry analyses the population of CD34.sup.+6.sup.+ HFSCs was significantly increased under this culturing condition employing a medium comprising EGF and the ROCK inhibitor Y27632 (termed Y medium; see FIG. 3).

[0432] An increased relative contribution of CD34.sup.+6.sup.+ HFSC population within the keratinocyte population in 3D-Matrigel ECM could be achieved by adding FGF-2 and/or VEGF (both mitogenic growth factors) in addition to EGF and the ROCK inhibitor Y27632 to the KGM culture medium (FIG. 2 and FIG. 3). In particular, culturing keratinocytes in 3D-Matrigel ECM and KGM medium comprising Y27632, EGF and FGF-2 (from here on termed YF culturing conditions; the respective medium is referred to as YF medium) or Y27632, EGF and VEGF (termed YV culturing conditions; the respective medium is referred to as YV medium) increased the relative contribution of the CD34.sup.+6.sup.+ HFSC population. However, the YF or YV culturing conditions did not increase the absolute numbers of CD34.sup.+6.sup.+ HFSCs, suggesting that both conditions rather promote HFSCs survival than HFSCs growth/expansion. By contrast, culturing cells in 3D-Matrigel ECM and KGM medium comprising Y27632, EGF, a FGF (e.g. FGF-2) and VEGF (referred from here on as 3C culturing conditions; the respective medium is referred to as 3 C medium) had a significant positive effect not only on the survival but also on the growth/expansion of CD34.sup.+6.sup.+ HFSCs within the cultured keratinocyte population. In particular, 6-fold increase in the percentage and 5-fold increase in absolute numbers was achieved by the 3C culturing conditions (FIG. 3).

[0433] To test whether EGF was required we compared Y and 3C conditions using a basal KGM medium with and without EGF. These culturing conditions employing 3D-Matrigel ECM and a KGM medium with the respective compositions are referred to as Y-E conditions and 3C-E conditions, respectively. The respective media are also referred to as Y-E medium and 3C-E medium, respectively. Interestingly, employing KGM without EGF did not affect the relative contribution of CD34.sup.+6.sup.+ HFSCs within the cultured keratinocyte population (FIG. 3A) but, the absolute numbers of HFSCs decreased by almost half (FIG. 3C). This indicated that although the growth of the CD34.sup.+6.sup.+ HFSCs within the cultured keratinocyte population is achieved in the absence of EGF, its presence supports and enhances the expansion of the CD34.sup.+6.sup.+ HFSCs, and therefore, EGF is a preferred component of the 3C medium that is employed under the 3C culturing conditions.

[0434] The presence of ROCK inhibitor (in the present example Y27632) in 3C conditions from the beginning of the culturing time is essential for HFSC growth and expansion (see above). To test whether ROCK inhibitor is indispensable for HFSC expansion in 3C conditions after the initial culturing phase (day 0 to day 4) we compared 3C medium to 3C medium without ROCK inhibitor (3C-Y) after cells had been cultured for either 2 and 4 days in 3C medium. These culturing conditions employing 3D-Matrigel ECM and a KGM medium with the respective compositions are referred to as 30-Y culturing conditions. The respective medium is referred to as 3C-Y medium. Interestingly, removing ROCK inhibitor after 2 or 4 days of culture did not affect the relative amounts of CD34.sup.+6.sup.+ HFSCs within the cultured cell population (FIG. 3E), but it decreased the absolute numbers of HFSCs (FIG. 3F). This indicated that although the growth of the CD34.sup.+6.sup.+ HFSCs within the cultured keratinocyte population can be achieved in the absence of ROCK inhibitor after 2 or 4 days of culture, its presence throughout the 14 days of culture supports and enhances the expansion of the CD34.sup.+6.sup.+ HFSCs, and therefore, a ROCK inhibitor and specifically Y27632 is a preferred component of the 3C medium that is preferably used during the complete culturing time.

[0435] Comparable growth/expansion of the CD34.sup.+6.sup.+ HFSCs within the cultured keratinocyte population was achieved when FGF-2 was replaced with FGF-7 or FGF-10 or FGF-18, and when VEGF (VEGF-164) was replaced with VEGF-121 in the 3C, YF and/or YV media (FIG. 4). This shows that different FGFs and/or VEGFs can be employed in the culturing method.

[0436] Furthermore, significant growth/expansion of the CD34.sup.+6.sup.+ HFSCs within the cultured keratinocyte population was achieved when the 3D basement membrane extract (Matrigel (Corning) or Culturex (Amsbio)) was replaced with 3D gels composed of collagen type I, collagen type I and laminin-332 or collagen type I and laminin-511, or collagen type I, Laminin-332 and Laminin-511 (FIG. 5). This shows that different 3D-ECMs can be employed in the context of the culturing method.

[0437] Flow cytometry analyses demonstrated that keratinocytes cultured in the 3C medium in 3D-Matrigel ECM also expressed the surface marker CD200 (FIG. 6) that marks stem/progenitor cells located in the bulge and hair germ compartments in human and mouse skin (Garza et al., 2011, J Clin Invest. 121(2): 613-622). While 19.91.5% of freshly isolated keratinocytes from P21 mice expressed CD200, 60.412.1% of the cells grown under 3C culturing conditions were CD200.sup.+ (3-fold increase). Similar to the fold enrichment of the CD34.sup.+6.sup.+ HFSCs in 3C cultures the absolute numbers of CD200.sup.+ cells were increased 5-fold (FIG. 6).

[0438] Cultures containing HFSCs grew into circular cell clusters, but in contrast to other 3D culture systems for epithelial cells known in the art (Lee et al, 2014, Cell 156: 440-455; Sato et al, 2011, Nature 469: 415-418; Sato et al, 2009, Nature 459: 262-265), they did not generate a lumen (FIG. 7). A preliminary EdU (5-ethynyl-2-deoxyuridine) incorporation experiment performed with cells from a single mouse revealed that both the CD34.sup.+6.sup.+ HFSCs and the CD34.sup.6.sup.+ non-HFSCs were actively cycling during a 12- to 14-day culture period under 3C conditions. Having this EdU (5-ethynyl-2-deoxyuridine) incorporation experiment repeated 7 more times confirmed the previous findings in what the CD34.sup.+6.sup.+ HFSCs are actively cycling during a 10-14-day culture period under 3C conditions (see FIGS. 8B and 8C). The in total 8 replicates (including the single experiment in FIG. 8A) indicate that 60.6% of these CD34.sup.+6.sup.+ HFSCs entered S-phase within 24 h (FIGS. 8B and 8C) whereas the CD34.sup.6.sup.+ (non-stem cell) population cycled slower (45.4%) entered S-phase within 24 h (see FIGS. 8B and 8C). Accordingly the EdU incorporation experiments confirmed that the CD34.sup.+6.sup.+ HFSCs divide, which is in accordance with the HFSCs being expanded during 3C culturing.

[0439] Immunofluorescence analysis of HFSCs cultured under 3C conditions revealed the presence of E-cadherin-containing cell-cell contacts, and the expression of additional HFSC markers including Keratin-15, CD34, and SOX9 (FIG. 9). In addition, the cultures were found to consist of only epithelial cells as 97.70.63% were positive for the epithelial cell adhesion molecule (EpCAM) (FIG. 10). RNA sequencing analysis of keratinocytes cultured in the 3C medium in the presence of 3D-Matrigel ECM confirmed that the global gene expression profile of these cells more closely resembled purified HFSCs than the total epidermal cells (see Principle component analyses (PCA) shown in FIGS. 11A and B and Heat map shown in FIG. 11C), supporting the fact that the 3C culturing conditions enrich for HFSCs. In addition to the RNA sequencing analysis, the upregulated expression of several HFSC identity genes (Cd34, Sox9, Tcf3, Id2, Wnt10a, Wnt7a) in cells grown in 3C culturing conditions compared to the cell mixture of origin (Epi d0) was confirmed by real-time qPCR in independent experiments (FIG. 11D).

Long-Term Culture of HFSCs

[0440] HFSCs cultured in 3D-ECM under 3C conditions were passaged every two weeks into fresh 3D-Matrigel ECM for a period of up to 32 weeks with no evident change in their potential to grow/expand and survive. Similarly, the percentage of CD34.sup.+6.sup.+ HFSCs within the cell mixture remained constant from the first passage onward (FIG. 12). Moreover, freeze-thaw experiments demonstrated that cultured HFSCs could be stored frozen and cultured again without evident loss of HFSCs or proliferative capacity (FIG. 13).

HFSCs Cultured Under 3C Conditions Retain their Proliferative Potential and Multipotency

[0441] To assess if cultured HFSCs from 3C cultures maintain their proliferative potential and multipotency, colony-forming assays that are the golden standard to quantitatively assess the proliferative potential of SCs (Jensen et al., 2010, Nat Protoc 5: 898-911) were performed. HFSCs (originating from cultivation under 3C conditions) plated on feeders at clonal density gave rise to more colonies that were also larger in size compared to freshly isolated keratinocytes containing 5.61.2% HFSCs (FIG. 14). Colonies derived from HFSCs that originated from 3C culturing conditions contained small, tightly packed, cobble stone-like colonies (FIG. 15) that are characteristic for holoclones observed in feeder-dependent 2D-cultures of HFSCs (Blanpain et al., 2004, Cell 118: 635-648; Greco et al., 2009, Cell Stem Cell 4: 155-169). This indicates that the 3C culture system enriches for HFSCs with high proliferative potential.

[0442] Full thickness skin reconstitution assays are used to evaluate self-renewal and multipotency of skin SCs (Blanpain et al., 2004, Cell 118: 635-648; Jensen et al., 2010, Nat Protoc 5: 898-911). In this assay, bona fide SCs will give rise to new HFs upon transplantation. HFSCs cultured under 3C conditions not only reconstituted the epidermis and produced hair, but were also more efficient in doing so compared to freshly isolated keratinocyte mixtures containing 5.61.2% CD34.sup.+6.sup.+ HFSCs (FIG. 16). Accordingly, this experiment demonstrates that the culture conditions used to expand HFSCs in vitro preserve their multipotency and capacity to self-renew.

HFSC Cultures can be Derived Both from Bona Fide HFSCs as Well as from Committed Progeny

[0443] To assess if presence of HFSCs in the initial cell mixture is required for the enrichment and expansion of HFSCs in culture, FACS-sorted HFSCs (CD34.sup.+6.sup.+) or committed epidermal cells (non-HFSCs; CD34 6.sup.+) from freshly isolated keratinocytes were used to establish HFSC cultures. Both cell populations were able to give rise to cultures containing CD34.sup.+6.sup.+ and CD34.sup.6.sup.+ under 3C culture conditions (FIG. 17A). To further validate these findings, HFSC were completely depleted from cultures by culturing total epidermal cells in 2D for 14 days (2D 14 d) and subsequently sub-culturing (passaging) these cells (non-HFSCs; CD34.sup.6.sup.+) in 3C conditions. After 14 days in 3C culture conditions HFSCs (CD34.sup.+6.sup.+) arose (FIG. 17B) from CD34.sup.6.sup.+, demonstrating that the 3C culturing conditions can induce generation of CD34.sup.+6.sup.+ cells from non-HFSC cells. Interestingly, regardless of the number of CD34.sup.+6.sup.+ cells in the initial cell mixtures, the cultures established an equilibrium of approximately 50:50 ratio of HFSCs and non-HFSCs (FIG. 18). This indicates that the 3C culturing conditions are able to induce generation of CD34.sup.+6.sup.+ cells from non-HFSC cells, and on the other hand, to promote differentiation of CD34.sup.+6.sup.+ HFSCs into non-HFSCs, establishing a balance between HFSCs and differentiated progeny in culture.

Addition of SHH Inhibitor to 3C Cultures Increases Proportion of CD34.sup.+6.sup.+ HFSCs

[0444] The observation that both CD34.sup.+6.sup.+ HFSCs and their committed progeny established a stable equilibrium in culture (FIG. 18) indicates that dynamic signaling crosstalk between these two populations occurs to establish and maintain this equilibrium under 3C culturing conditions. Surprisingly, further addition of the SHH inhibitor cyclopamine to HFSC 3C cultures led to an even elevated percentage of CD34.sup.+6.sup.+ HFSCs concomitant with a decrease in total cell numbers indicating that SHH regulates the balance between HFSCs and the differentiated progeny (FIG. 19). Accordingly, the addition of SHH can be used to further increase the proportion of CD34.sup.+6.sup.+ HFSCs in 3C HFSC cultures.

Addition of BMP Inhibitors Increases Cell Proliferation and Decreases the CD34.sup.+6.sup.+ HFSCs Population

[0445] To evaluate the effect of BMP inhibitors on cultured HFSCs, HFSCs were cultured under 3C conditions in the additional presence of the BMP inhibitors dorsomorphin or K02288. Both BMP inhibitors caused a decrease on the percentage of CD34.sup.+6.sup.+ HFSCs (FIG. 20). Accordingly, addition of BMP inhibitors counteracts the expansion and enrichment of CD34.sup.+6.sup.+ HFSCs observed under 3C conditions.

EXAMPLE 2: IN VITRO CULTURE CONDITIONS FOR EXPANSION AND MAINTENANCE OF SKIN CANCER STEM CELLS

Materials and Methods

Mice

[0446] Mice of an age between 4 and 40 weeks were used for experiments. Tumor cells were harvested from Tg(Krt14-rtTA)Tg(tetO-KRas2) mice (Fisher et al, 2001, Genes Dev 15: 3249-3262; Nguyen et al, 2006, Cell 127: 171-183), where overexpression of mutant KRas was induced by feeding the mice with doxycycline-containing chow (1 g/kg), leading to formation of benign papillomas and squamous cell carcinomas.

[0447] Animals were housed and maintained according to FELASA guidelines at the animal facility of the Max Planck Institute for Biology of Ageing, Cologne, Germany. All experiments were approved by local authorities.

Cell Isolation and Culture

[0448] Tumor cells were isolated by mincing tumor biopsies and incubating them in 0.25% collagenase (Sigma), 62.5 U/mL DNaseI (Roche) in HBSS (Hank's Balanced Salt Solution, Gibco) for 60 min with gentle agitation at 37 C. Cell suspensions were filtered through 45 m strainers and centrifuged at 300g for 10 min. Single cell suspensions were used for HFSC cultures as describe above. The cells were cultured in identical conditions as described in Example 1 for HFSCs.

[0449] Additional materials and methods are described in Example 1.

Results

Development of Culture Conditions for Expansion and Maintenance of Epidermis-Derived CSCs

[0450] The main challenge to study and manipulate cancer stem cells (CSCs) is the difficulty to obtain these cells in sufficient quantities and the lack of an amenable culture system maintaining the unique cellular properties of these cells. As the culture conditions described above successfully supported maintenance/survival and/or expansion/growth of HFSCs, it was next addressed whether similar conditions could also be used for culturing other epidermis-derived stem cells. In particular, the culturing of epidermis-derived CSC was tested. CSCs from cutaneous papillomas have previously been described to lack linage (Lin) markers (CD140a, CD31, CD45) and to express EpCAM and the markers CD34 and 6 (also expressed by HFSCs). These CSCs account for 20-30% of the total cells in benign papillomas in mouse models (Lapouge et al., 2012, EMBOJ 31: 4563-4575). In accordance with previous studies, FACS analyses indicated presence of 2216.6% EpCAM.sup.+CD34.sup.+6.sup.+ cells (hereafter CSCs) in the pool of tumor cells isolated from papillomas obtained from the Tg(Krt14-rtTA)Tg(tetO-KRas2) mouse model (FIG. 21). Culturing these tumor cells in 2D with standard keratinocyte growth conditions completely depleted the EpCAM.sup.+CD34.sup.+6.sup.+ population. Culturing these tumor cells under the 3C culturing conditions as described above in Example 1 (e.g. in the presence of Matrigel, Y27632, EGF, an FGF (here FGF-2) and a VEGF) resulted in 972% of EpCAM.sup.+ cells, but the EpCAM.sup.+CD34.sup.+6.sup.+ population was reduced (6.12.5%) after a 2-week culture period (FIG. 22). However, the 3C culturing conditions supported cell growth for more than 5 passages after which the percentage of EpCAM.sup.+CD34.sup.+6.sup.+ cells had reached 69.3% concomitant with a 6.4-fold increase in absolute cell numbers compared to the number of cells seeded at the beginning of the cultures (FIG. 23). These cultured CSCs grew in clusters similar to the HFSCs (FIG. 24), and could be passaged up to 14 weeks without a significant loss of their growth potential. Moreover, freeze-thaw experiments demonstrated that cultured EpCAM.sup.+CD34.sup.+6.sup.+ cells could be stored frozen and cultured again without evident loss of EpCAM.sup.+CD34.sup.+6.sup.+ cell numbers.